You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, the GSI soil analysis (DO_GSISOILDA) calls regridding code from GDASApp/DA-utils. This regridding is called from the sfcanl and esfc jobs. This introduces 2 complications:
Requires GDASApp to be built to do the soil DA, but only requires a tiny executable.
On hear at least, requires that the modules from ufsda be loaded in place of the standard workflow modules for these jobs.
specifically jobs/rocoto/sfcanl.sh and esfc.sh
must call load_ufsda_modules.sh instead of load_fv3gfs_modules.sh.
We need a more general solution that can call the same module script with / without the soil DA.
What should have happened?
It would be preferable to be able to run the GSI soil DA without compiling GDASApp.
Need to remove current manual switch of the module file in the job scripts when using the soil DA.
What is wrong?
Currently, the GSI soil analysis (DO_GSISOILDA) calls regridding code from GDASApp/DA-utils. This regridding is called from the sfcanl and esfc jobs. This introduces 2 complications:
specifically jobs/rocoto/sfcanl.sh and esfc.sh
must call load_ufsda_modules.sh instead of load_fv3gfs_modules.sh.
We need a more general solution that can call the same module script with / without the soil DA.
What should have happened?
It would be preferable to be able to run the GSI soil DA without compiling GDASApp.
Need to remove current manual switch of the module file in the job scripts when using the soil DA.
What machines are impacted?
All or N/A
What global-workflow hash are you using?
Issue introduced for PR #3295
Steps to reproduce
setting DO_GSISOILDA to YES will crash in the sfcanal and escf jobs if the module script is not manually switched to load_ufsda_modules.sh
Additional information
No response
Do you have a proposed solution?
Best solution is probably to move the regridding code. @CoryMartin-NOAA
The text was updated successfully, but these errors were encountered: